Study design and setting

This is a multicenter retrospective study including children with NMD diagnosed with SDB by PSG and recommended home NIV initiation from 2010 to 2023 in three Canadian centers: the Stollery Children’s Hospital (Edmonton, Alberta), the Alberta Children’s Hospital (Calgary, Alberta), and the Hospital for Sick Children (Toronto, Ontario). These tertiary care pediatric hospitals operate outpatient NIV programs and sleep laboratories, incorporating pediatric PSG for the initial diagnosis of SDB as well as titration studies to determine the optimal NIV settings prescription. This study was approved by the Research Ethics Boards (REB) at the Stollery Children’s Hospital, University of Alberta, Edmonton, Alberta, Canada (REB 00122183), the Hospital for Sick Children, University of Toronto, Toronto, Ontario, Canada (REB 1000080153), and the Alberta Children’s Hospital, University of Calgary, Calgary, Alberta, Canada (REB 00122183/pSite-22–0035).

Participants

Children and youth (0 to 18 years) were included if they had (1) a confirmed diagnosis of NMD [1, 3], (2) a diagnostic PSG with reported SDB by interpreting physician, and (3) documented recommendation for NIV initiation, NIV prescription, or referral to an NIV program. Eligible participants were identified through the patient lists from NMD clinics and crossmatched with sleep laboratory records and patient lists from NIV programs. Patients who were deceased, received a tracheostomy, those transferred to a different center, or patients who declined prescription or referral for NIV initiation were excluded.

Data collection

Data was collected from the electronic health records at each site and stored using a REDCap (Vanderbilt, USA) database [19]. We reviewed medical records of eligible participants and collected demographics—age, sex, ethnicity, body mass index (BMI), and percentiles for children and youth [20], clinical data including underlying disease and other comorbidities. Comorbidities were defined as any additional chronic diagnoses, other than their underlying NMD [21], documented in the medical charts. These were classified based on the affected body functions and structures into the following categories: orthopedic, respiratory, cardiac, psychiatric, and gastrointestinal. We collected diagnostic PSG results prior to NIV initiation. PSG data included AHI, obstructive AHI (OAHI), central AHI index (CAHI), oxygen saturation index (SpO2) by pulse oximeter, etCO2, and tcCO2, both during REM and non-REM sleep stages. Diagnostic PSGs were requested based on symptom assessment by treating clinicians and were scored and interpreted according updated AASM standards [12].

Outcome measures

Outcome measures were collected from available medical records at the follow-up visits within 6 to 12 months from NIV initiation including:

Changes in lung function: Forced vital capacity (FVC), forced expiratory volume in 1 s (FEV1), peak cough flow (PCF), maximum inspiratory pressure (MIP), and maximum expiratory pressure (MEP) following the American Thoracic Society (ATS) standards [22] at two time points: (1) the initial NIV clinic visit (baseline) and (2) the NIV clinic follow-up visit. 

Reported patient’s clinical improvements documented in medical records: Clinician’s documentation in medical charts of improvements in sleep quality (more sleep hours, less nocturnal awakenings, more rested in the morning), breathing during sleep (less snoring, less pauses in breathing, less work of breathing), daytime function (more energy, more alert, less daytime sleepiness, less fatigue), improvement in behavior or mood (less emotional or irritable, more interactive, easier parenting), learning, attention or school performance (more focused on school, able to do homework, teachers’ comments on improvements) and overall health (fewer respiratory infections, less sick, fewer hospital admissions and visits to the emergency). This information was recorded by clinicians in participating sites based on child and parental reports during routine visits.

NIV adherence data from machine downloads: NIV adherence data were collected from secure digital cards and/or cloud-based datasets and included average hours of NIV use per night and percentage of days with NIV use > 4 h within a 30-day period close to the regular follow-up visits within 6 to 12 months from NIV initiation.

NIV-related reported complications in medical records: Mask-related issues (feeling of suffocation, interface/mouth leak), child sleep disruption (not falling asleep with mask on, mask displacement/removal during sleep), other (non-specified).

Data analysis

Descriptive statistics were used to summarize the data. Data distribution was assessed using the Shapiro–Wilk test. Continuous variables were reported as median and interquartile range (IQR), and categorical variables were presented as frequencies and percentages (n, %). Comparisons between continuous variables were performed using the Mann–Whitney U test. The Kruskal–Wallis test was used to compare continuous variables across the groups. Differences between categorical variables were assessed using Pearson chi-square or Fisher’s exact test, as appropriate. Bonferroni-corrected post hoc tests were applied for multiple comparisons.

For the purpose of this study, we classified children in three groups considering (1) CHEST guideline criteria [8], (2) authors’ proposed REM-related criteria based on known pathophysiology of SDB in REM sleep in children with NMD [3, 5, 7], and (3) AASM-based criteria for moderate/severe OSA and nocturnal hypoventilation [12, 23] (Table 1).

Table 1 Sleep-disordered breathing criteria for indication of non-invasive ventilation

Agreement between CHEST guideline, author’s-proposed REM-related, and AASM-based criteria for AHI, nocturnal hypoventilation, and nocturnal hypoxemia was evaluated using Kappa statistics, and Kappa level of agreement was classified according to McHugh [24].

Analyses of NIV adherence and clinical outcomes between groups were performed in children with data available within 6–12 months from NIV initiation to avoid potential confounding during the period of acclimatization within the first 6 months after NIV initiation. Wilcoxon signed-rank test was used to check for differences in baseline pulmonary function and after NIV initiation between groups. Missing follow-up data were assumed to be missing at random. Sensitivity analyses were performed to assess the potential impact of missing follow-up data.

Hypotheses were 2-sided and considered statistically significant if p < 0.05. Data analyses were performed using SPSS, version 29.0 (IBM Corp., Armonk, NY, USA).